Air gap, magnet shape, and voltage questions

So I'll apologize in advance for the long post with lots of questions...
Sorry.

I'm running a Heinzmann PMS126 axial flux motor in my Electrathon car.
It operates on 36v due to my battery weight restrictions although the controller (an ACD 4805) can supposedly handle up to 60v and according to the spec sheet the motor operates from 48v-560v (crazy).
So I'm undervolting a bit.
I built a hydraulic load cell a few years ago for motor testing and although it is not a dyno (putting out hp numbers) it does allow me to determine if a change I've made in voltage, timing, or controller settings, etc. helps or hurts efficiency by reproducing pump speed and output pressure and comparing wattage usage figures.
I generally test at a speed and pressure that equates to about 1/4 of the max of what it sees in competition.
I decided to peek under the hood of the motor after 3 years of running it stock and was surprised to see the left and right air gaps were uneven. .035" on one side and .055" on the other. Since everything I've read says that with air gaps "less is more" I decided to even them up and set them both at .025". I then ran it on my load tester and compared it to a baseline test performed before I reshimmed expecting to see an efficiency improvement.
Instead I saw an INCREASE in power usage of approximately 8%.
Confused, but willing to follow the numbers, I reshimmed again setting the air gaps to .070" and ran it again.
This time I saw an efficiency improvement over the stock setup of about 6%.
So what happened here?
My thoughts are that since the motor is designed to operate at a higher voltage than it does currently, that the flux created by the stator windings on 36v isn't sufficient to fully overcome the attraction of the rotor magnets to the stator laminations at such a narrow air gap.
Kinda like how you can get two opposing magnets to stick to the same piece of metal after you overcome the initial repulsion.
By increasing the air gap, I have moved the rotor far enough from the stator for the windings to be able to repel the rotor magnets more efficiently.

Is that right?


Another thing is the shape of the magnets in the rotor is not symmetrical.
I assume it has something to do with reducing cogging torque but perhaps not.
Anyway, does this mean that the motor should be more efficient in one direction vs the other?

And lastly...
I've always heard that motors are more efficient at higher voltages rather than lower ones as it reduces the resistance losses (or something).
So I decided to test my stock motor (pre-re-shimming) at 36v, 48v and 60v.
Oddly, the wattage requirement actually went up slightly with voltage. About 5% for each voltage jump at 1/4 max load.
Is this a function of the controller being less efficient while switching higher voltages or something else?

Racing Electrathon means always looking for more efficiency...

Thanks for your time.

Are you sure that 560v is not a typo? Maybe 56v? Who even has 560v?

Anyway, air gap is a give and take thing, theres no 'right one'. You can make it whatever you want, depending on
what you want the motor to do. If you reduce it, you get more power or torque, but you get more heat, and less
speed. If you increase it you get less torque, less heat, and more speed. Of course, as usual, there are more things
involved, but thats the basics.

You simply have to experiment with it on your motor, like you've been doing, :thumb: until you have it dialed in. The air gap is
probably the most sensitive adjustment you can make on a motor, but theres no free lunch here, everything else changes with it.

Heres a simple chart on the directions to go on motor tweaks, I'm sure you know this stuff, but others may not, at any rate,
it doesn't hurt to post info.





And a nice thread on the same info. that I've found useful; http://web.mit.edu/first/scooter/motormath.pdf

As far as the shape of your magnets, I haven't seen that particular arrangement, and must be a proprietary thing that
they have discovered to work better.

Hope some of this helps.

APL said:

...
And a nice thread on the same info. that I've found useful; http://web.mit.edu/first/scooter/motormath.pdf
...

Click to expand...

Use information from that link with caution. I question the accuracy.

Regards,

major

The motor should have a KV rating so for every volt applied the rpm will rise by that given amount, to operate at 36v to 560v is not going to happen efficiency at 36v would be horrible there's to wide a design voltage, For example if the KV is 14 per volt then at 36v it would have 500rpm and at 560v it would have 8000rpm.

When we run a motor from a higher voltage the no load current will increase slightly, it's to do with the i2r losses because it's resistance is fixed when the voltage is increased the current draw will rise too v÷r=i but in a motor it becomes more complex with induction so the formula is longer but that's the just of it, to reach its max efficiency you must use the motors recommended max volts and load it so it can reach its max rpm if a motor is undervolted and has a efficiency of 80% it will reduce then if the rpm is kept low again it will never reach efficiency.

I've never run an axial motor before but I do know the smaller the air gap the greater the induction and cogging of the motor so it's got more current going in but the torque output should have increased per watt applied aslong as theres no physical contact in the motor increasing friction losses, windage should be reduced to but it's less of an issue on a small motor with fairly low rpm.

I looked at the Heinzmann spec sheet and got the impression that there was a different winding available for 560V.

I have to assume that the speed is limited by the controller as it doesn't change proportionally with voltage on the chart.

Even so, no one has an opinion on why efficiency would increase with a larger air gap?

There you go. Different voltage models have different voltage and torque constants, so obviously are wound differently.

About air gap and efficiency, I'd need to see a lot more details. For me, there are too many unknowns about load and measurements, as well as the motor design and how the modifications were done.

Generally speaking, small change in air gap doesn't significantly or consistently affect the efficiency.

Regards,

major

major said:

There you go. Different voltage models have different voltage and torque constants, so obviously are wound differently.

Generally speaking, small change in air gap doesn't significantly or consistently affect the efficiency.

Regards,

major

Click to expand...

I didn't see anything on the sheet to indicate that the different voltages listed were indicative of different models...

If magnetic force across an air gap is inversely proportional to the distance squared (which I understand it is) then wouldn't doubling it from .035" to .070" not be considered a "small change"?

On the chart which archer321 posted, I see 48V, 80V, 96V, 330V, & 560V models listed each having different different design constants.

With the "air gap" in permanent magnet motors you need to consider the distance of the magnet. Here is some discussion on the subject. https://www.eng-tips.com/viewthread.cfm?qid=98210

Edit: Another reference.

The air gap referred to here includes the length of both the physical air gap and the actual radial thickness of the magnet. The implied assumption here is that most magnet materials used in DC motors have a permeability equal to that of air.

Click to expand...

From:
http://www.consult-g2.com/papers/paper5/paper.html

Increasing the physical gap between the magnet surface and the armature steel will increase flux leakage having the effect of field weakening. The amount of field weakening which I imagine resulting from that change is unlikely to significantly affect overall machine efficiency, IMO. There will also be other effects like changes in armature reaction and inductance. It is difficult to say what the resulting efficiency would be. I'd think you'd be better off looking at a good simulation program than trying to quantify it with a hydraulic pump. But at least you're trying. Good for you.

Regards,

major

Presumably you have to eliminate from testing any possible effect the change in air gap might have on the operation of the controller driving the motor?

With a smaller gap, the flux in the iron will be higher, resulting in higher iron losses (and more torque). Higher flux also changes the kV of the motor, so the optimal gearing might change.

Imagine just spinning the motor with no power applied. If you remove the magnets, it will spin very easily. The stronger the magnets, the more drag there is.

If you want maximum power/torque density, keep the gap as small as possible. If you want maximum efficiency, the gap may need to be a little bigger.

Your test setup is the best way to optimize the setting.

anything wrong with adopting the term 'flux-gap' instead?
whether the motor is operating undersea or vacuum of outer-space it's the same gap

Simple models suggest that a motor is most efficient when its copper losses and iron losses are equal. Roughly, copper losses are proportional to square of current while iron losses are proportional to square of speed.

I'm not exactly sure how your hydraulic load cell works, but perhaps you are running your motor at an operating point where iron losses are larger than copper losses? Increasing air gap increases the speed constant Kv and decreases torque constant Kt. This means for the same torque load, your motor will be using more current which moves your motor towards a more optimal operating point?

If you have a fixed load, you can just design your motor and gearing to maximize efficiency at that one operating point. If you have a variable load, then max efficiency is a difficult target. Actively changing gearing (common) or air gap (less common) can help you keep your motor at an optimal operating point at the expense of weight and cost that might be better spent elsewhere (many argue that the motor itself is the ideal place to direct resources).